How Brands Win "Machine Recommendations" in the AI Era

At the recent Shenzhen SEO Conference, Jamie I.F. gave a thought-provoking talk.

About the speaker: Jamie I.F.

  • Founder of AffiliateFinder.ai: an AI tool that helps brands find influencers and affiliates who are already promoting their competitors, and discover top creators in niche markets for recruitment.
  • Founder of Endorsely.com: a software platform offering free affiliate marketing tracking for SaaS brands.

His core message was striking: in the age of AI, Chinese brands that want to win American customers must redefine their marketing strategy — actively influencing large language models (LLMs) like ChatGPT to become their AI’s first-choice recommendation, and thereby dominate future market competition.

The talk revealed how AI is reshaping users’ decision-making paths and gave going-global brands a clear action guide. This article provides an in-depth, objective breakdown of the talk’s core content.

1. The Future Revenue Mainstream: Recommendations from AI

The talk opened with an irreversible trend: users are shifting from “actively searching” to “asking AI.”

The data shows 700 million people use ChatGPT every week, projected to reach 1 billion by the end of 2025. Millions of users ask AI for shopping advice daily and trust its recommendations highly. Even more critically, traffic driven by ChatGPT recommendations converts at up to 10x the rate of traditional SEO or paid traffic.

But behind this efficient conversion lurks a “winner-takes-all” Matthew effect.

“Winner-Takes-All” in the AI Era

When LLMs give recommendations, they’re inherently risk-averse. They tend to recommend brands that are already widely known and have lots of positive information across the internet — rather than discovering niche or emerging brands.

This mechanism creates a self-reinforcing cycle:

A well-known brand gets mentioned frequently → the LLM deepens its “awareness” of it and recommends it more → the brand earns more sales, reviews, and discussion → that new content gets used to train the next generation of AI models → eventually, the brand becomes the “default answer” in its category.

The speaker warned that once this loop closes, it becomes extremely difficult for new brands to break in. So brands must act immediately, using the next 5 years of AI search development to position themselves as the go-to choice in their industry — or risk being marginalized.

2. The Two Core Pathways to Influencing AI Recommendations

So how can brands influence LLMs and make them “fall in love”? The talk pointed to two key pathways, emphasizing the strategic importance of affiliate marketing along the way.

Pathway 1: Influence AI’s Information Retrieval

When AI tools like ChatGPT and Perplexity answer user questions, they retrieve the web in real time (mostly Google and Bing search results) and synthesize the top-ranking information into an answer. Google’s AI Overviews follow similar logic.

Action plan: your strategy should be getting your product or service into as many high-ranking articles as possible — especially listicle content like “Top 10” and “Best of.” By setting up an affiliate program, you can effectively incentivize the operators of those high-ranking sites (who mostly monetize via affiliate commissions) to recommend your brand in their content.

Pathway 2: Influence AI’s Future Training Data

An LLM’s knowledge base comes from its massive training data — crawls of the entire internet, including blogs, YouTube videos, Reddit threads, social media, and more. AI forms its “knowledge graph” and recommendation tendencies by analyzing how frequently and in what context a brand appears across that sea of data.

Action plan: brands need to recruit affiliates and influencers at scale to create content across all major platforms. This content not only drives sales directly — more profoundly, it will be learned and absorbed by future AI models (like GPT-6), effectively “implanting” your brand into AI’s long-term memory and making it the authoritative answer in your niche.

3. Four Concrete Strategies to Boost AI Recommendation Rankings

Building on the two pathways, the talk proposed four specific, executable strategies that work as a combined playbook.

Strategy 1: Build Topical Authority with the RRF Algorithm

The talk revealed an important finding: AI models like ChatGPT may be using an algorithm called “Reciprocal Rank Fusion (RRF)” to integrate search results.

The RRF algorithm doesn’t just value rankings for a single core keyword — it weighs a site’s overall performance across a constellation of related long-tail keywords. When AI handles a complex query (like “best CNC machining service in China”), it simultaneously searches multiple related questions in the background (like CNC service reviews, small batch CNC China, etc.).

If your site provides lots of high-quality content for that whole constellation of questions, then even if no single page ranks #1, your overall RRF score will be higher thanks to being “comprehensive” and “scoring on multiple fronts” — making you more likely to get the final AI recommendation.

How to execute: build content clusters around your core business. Deeply research every question users might have before, during, and after purchase, and create professional, in-depth content for each. When your site becomes the “knowledge base” for a niche, AI naturally treats you as the expert in that field.

Strategy 2: Make Good Use of “Parasite SEO”

Building topical authority takes time. As a complement, the talk recommended “parasite SEO” as a quick, low-effort tactic.

The core idea is to leverage high-authority, high-trust platforms like Reddit, Quora, LinkedIn, and Medium to publish content related to your brand. Since search engine algorithms currently favor these authoritative “hosts,” the same content published on these platforms often ranks faster than on a brand’s own new website.

Key points:

  • Find the “host”: look for hot topics or communities on these platforms related to your product.
  • Value first: share experiences and solve problems as a real user, rather than pitching directly. For example, in a thread discussing a competitor, you can objectively share your usage experience and “casually” mention your brand as an alternative — it comes across as far more credible.
  • Plant your info: this way, you embed your brand information into the high-quality “corpus” that AI models prioritize learning from, indirectly influencing AI’s perception.

Strategy 3: Redefine the Strategic Role of Affiliate Marketing

The talk stressed that brands must elevate affiliate marketing from a mere “sales channel” to “a strategic tool for influencing AI.” At its core, it’s a crowdsourced brand advocacy model — incentivizing hundreds or thousands of external creators to continuously speak up for your brand in every corner where AI is watching.

Strategy 4: Proactively Embrace AI Overviews

For Google AI Overviews, now fully rolled out in markets like the US, brands should actively embrace them. Getting cited in AI Overviews brings massive brand exposure and a trust endorsement.

There’s no shortcut to earning citations — it still comes down to SEO best practices:

  • Create high-quality, helpful, people-first content.
  • Clearly answer users’ specific questions within your content.
  • Use structured data (Schema Markup) to help Google better understand your page content.

Summary: Brand Is the Only Moat in the AI Era

The talk’s core argument can be summed up as: independent site operations need to shift from a “traffic mindset” to an “influence mindset.”

Clicks are no longer the only metric that matters. Your brand’s “presence” and “authority” in the AI world are what determine success or failure. Brands need to “infect” AI’s knowledge base from every angle — content building, community penetration, affiliate partnerships — until AI willingly endorses your brand.

In an era when AI is rewriting the rules of information distribution, no brand influence means no future.

References: